I am planning to buy a dedicated server and a shared server from a hosting company.
Basically I want to have a redundant server so that if one file server goes down, there is no downtime.
Somehow the servers would need to be constantly synchronized so that the files saved to one are immediately saved to the other server as well.
Can somebody told me how to setup the 2 server so that my dedicated server can serve as the main server and if the dedicated server is down, the shared server can automatically be activated and visitors of my website will auto be redirect to the shared server.
Do i need to setup any backup DNS too so that when the dedicated server is down, it will auto redirect the user to the shared server.
I have two questions that hopefully someone will be able to help me out with. The first, is my partner and I want to provide server redundancy for our clients. Our set-up is as follows: two identical servers with multiple virtual machines (about 4) on each. One for SQL, one for IIS, etc. We also have a ServerIron XL to connect the two together and provide replication and load balancing. So the question is, has anyone here used a ServerIron XL and how easy is it to use/how effective is it? The second question is, we would like to provide our clients with hosted Exchange. I have set up and maintained a few Exchange 2007 servers, but only for a single company with a single domain. What would be the best way to go about providing a hosted Exchange solution?
There are several methods of offering network path redundancy. The basic decision for me has come down to:
Do it at Layer 2 Do it at Layer 3
At the moment we have layer 2 redundancy to each server. 2NICs on the server up with one having the IPs for apache/mysql/etc. each nic going into a seperate switch and the switches connected together, with 2 routers running VRRP to handle the gateway. Everything is Vlaned.
So basicaly the switch redundancy is done by spanning tree and the IP redundancy is done with a process on the router/server to move the IP to the other router/server NIC if there is an issue.
I am thinking about going with 2 fully seperate networks. in differant subnets. where each router would have a gateway. the routers would talk to each other and they would speak OSPF or ISIS with the servers. This way I would move hosting onto loopback IPs on the servers and those loopback IPs would be advertised to the both routers through seperate networks. Again everything would be Vlanned.
This has the advantage of getting rid of spanning tree which has caused issues from time to time. It would keep redundancy up and we could standardize on the routing daemon ran on the servers allowing all of the various OSs we run to have the same basic config for network redundancy. Where now each OS tends to have its own solution. This would make life easier from a config and troubleshooting point of view.
what happens when I have 2 DNS servers on my domain and 1 fails? around 50% of the access fails... or they detect DNS timeout and try the other one, so 50% of the access would just take more time?
after months of disruption moving servers into a new data centre, our once reliable colocation company has now had nearly 6 hours downtime in the last 16 hours. So much for network redundancy.
Basically have 2 hosting accounts at different providers...each set-up for the same domain name...and then somehow wtih DNS make it so if host #1 goes down traffic goes to host #2 (which would basically be a splash screen explaining that host #1 is down and will be back soon).
DNS isn't my strong point, but I do know you can do this with MX records...so if the first server fails it tries the next until it gets a working one or reaches the end of the list. I'd just like to do it is A records.
It wouldn't be as simple as setting the nameservers like this would it?
Would it use the host1 nameservers as long as they're online, and if not failover to the host2 nameservers? If so, great, but what if the host1 nameservers are online but the server itself is not.
What are the smaller shops doing for switch redundancy? We have all our machines on dual Com Ed feeds but most switches in the $1k-$3k range only have one power supply. We recently had a power strip go flakey and of course the switch was plugged into it.
Is the best solution getting two switches and hooking each machine up to both? How hard is that to setup in Linux? I've used keepalived for whole machine failover but not for network failover.
I host several web clients that were recently impacted by the crap at ThePlanet. As I think about how to be more redundant (and repetitive) I'm not sure of my options.
What's the best practice to ensure that if you have a server at a data center that goes out, that you can (somewhat) easily switch over to a different server? I suppose one solution is to have 2 servers at 2 physical locations, and then you could just change the DNS record in the event of failure, but is there another solution I'm not aware of? Is there a good resource I can goto to read up on this info?
I am working on setting up a few servers that run cPanel and the usual software, minus any sort of DNS server. These servers are setup to sync all DNS records over to a cPanel "Dns Only" (ns1.domain.com), which in turn syncs all its records to a backup DNS server (ns2.domain.com).
In other words: [Domain registrar] | | [NS1]--Sync-->[NS2] | (Sync to NS1) | Cpanel Web Server
Now, say something happens to NS1 and the sever goes completely offline (i.e. power supply dies, CPU goes bad, etc.), which of the following scenarios would actually happen:
1) Because the registrar lists both NS1 and NS2 as NS records, NS1 would time out and the DNS lookup would look to the secondary DNS (NS2) for the record.
2) The registrar would randomly give out NS1 OR NS2 because of round robin, and if NS1 is given to a client as the result of a DNS lookup, the site will appear down, however if the client happens to be referred to NS2, the site would appear online.
3) The site would be down no matter what.
So, if someone with knowledge on the subject wouldn't mind enlightening me as to which of those would actually take place in the event of a failure on NS1, and maybe some suggestions as to keeping the DNS truely "redundant", then I would greatly appreciate it.
And Im aware that there are many 3rd party services that will take care of the DNS records and make them redundant (DYNDNS, DNSMadeEasy, etc) But I would prefer to keep the DNS in our full control.
Myname.com is a very popular website and his content mirrors in 2servers (Server A and Server B) in 2 different datacenters (exactly same content).
When Server A fails due to datacenter or rack problem and no site is responding, then I need some service to transfer ALL traffic to server B. Is there any (paid or free) service doing such a thing? If yes, what about dns propagation? As far as I am concerned, dns propagation is a very slow procedure depending on each internet provider.
So how could we instantly make Server B to come live to ALL visitors? Is there any quick and reliable dns redundancy system without having to wait for dns propagation?
I have a mysql based e-comm site that works only with older version of php and mysql. My shared host does not meet this requirements, so I have to go with a dedicated or vps.
I am looking for a server to have a fast mysql query.
Is there any vps providers that offer redundancy of the server, meaning if the server crash there should be a standby sever to take over?
You always hear of stories of a company that had such overnight success that they servers couldn't cope at first. Of course, this is rare, although we'd all wish for it. But how can you stand ready for these situations without investing in big equipment that might not even be needed at the end.
Knowing that as a start up, you're low on cash. But if you hit the tree line upon takeoff, there might not be second chance.
I've got a decent server, quad processor, 12GB ram, terabyte of hot swappable RAID 10 disk. Its at an Equinix site. They provide UPS, backup generators, etc and fiber to the usual major backbone folks.
But its a lights out facility. I want someone else to handle backups, rebooting when needed, etc.
I'd like to let someone else provide the service. Use my 2U server or I can rent yours.
Need MySql, Apache, Tomcat, SSL, Java, ssh access. I've got wildcard certs for SSL, domain registered, etc.
I'm guessing that I can get this for about $500 per month with low bandwidth (this is an IT application, not peer-to-peer or torrents.). Let me know if my budget is off base. I think one megabyte/second sustained over the month, 95% averaged is suffiicent. At least until the business grows, we can then talk about getting racks of blades at higher cost.
I don't have a problem with Bluehost, but after getting this email tonight I would have thought that it's not that hard to have enough fuel and a big enough generator to last more than five minutes after a power outage. It seems like pretty poor redundancy planning if a major host can't last more then five minutes with no power. Hospitals and other places don't have any problem doing it with similar or bigger power demands.
Dear Bluehost Customer,
This evening (July 14th) from about 5:25pm-6:55pm many of our servers were offline causing significant downtime for many of our users. The outage was due to a severe power outage in the north end of Orem, Utah where our servers are located. We do have UPS backup as well as diesel generators, but at about 5:30 they finally gave out. The power outage was for much longer than that period of time, but the reserve power was eventually consumed in its entirety. When it rains it pours.
For users on box65-box145 there have been periodic problems with the Redhat linux kernel that we were using that was causing problems with the filesystem that your data is stored on. This issue has been causing periodic problems for users on those boxes. In the last few days we have resolved that issue which also caused those boxes to require a reboot.
The downtime is extremely regretable. We apologize profusely for the inconvenience to our customers and in turn to those who were trying to visit your sites during the outage. With the fixes we have put in place in the last few days coupled with other upgrades you should experience MUCH better uptime in the future.
Considering skipping VPS and going to a colo setup for a handful of sites. Nothing major, so the server will be very entry level, but with redundancy in mind (software RAID1 and 2 nics). But I have a few basic questions:
How good is hot swapping in Linux? This was very hard to me to find out online. I am getting a 1U rack with a hot swap backplane and 2 SATA drives. I won't be using any commercial software with my setup.
How does redundant NIC work? This is new to me and am wondering how this is setup.
I think I can shop around NYC for a 1U slot for around $40 a month. I don't need a lot of transfer, but would like a decent pipe. The thought of 1Mbit sounds unattractive (transfer is around 100KBytes/s, right?). How much would 10Mbit cost? I found some quotes but they seem way too much (I could be wrong).
Not sure where this belonged so put it in web hosting as it could cover dedicated, colo or shared.
In a couple of months I am going to be launching a new e-commerce website for the company I work for, so it is absolutely vital that the website never goes offline, otherwise I will be in deep ****.
It will be hosted on a server that has been built to be very redundant; 2x Hotswappable Redundant PSU, 4x Hotswappable HDDs (RAID 6), Redundant NIC.
Sods law though to achieve this redundancy it will need to be on a colocated server, so if there is a major problem like the motherboard dying, then I will have to get down to the DC (3 hours away) and replace the entire server. Which wouldn't be practical as we would need a replacement server, all setup and ready to go, to do that.
Also, if there is a problem with the data centre itself, then we would be stuck until it is resolved. I have hosted sites with data centres in the past that experienced power or network problems which sent the sites down and it was totally out of our control to either prevent the downtime or restore the sites.
So to increase redundancy, and minimise downtime, I was thinking of load balancing between two servers. But then I assume they would both have to be hosted in the same data centre.
I was also thinking of having the site hosted on the server, but having some shared hosting (or a cheap dedi) ready to go just to fall back on if the server goes down. That way I can forward requests to the shared hosting as a temporary measure whilst the main server is restored. But the problem with that is I would probably need to get another SSL certificate. I would need to get the payment gateway provider (protx) to change the settings for the IP of the shared hosting, which knowing them will take 72 hours to process.
Budget really isn't an issue here if it can be justified, just looking for some ideas at this stage. There is no way this site can be down for a second longer than it needs to be.
I just want to use a server for file sharing, it will have nginx and that's it. I'm looking at centos, or freebsd, but I been using centos forever now and I'm not sure how to use freebsd, should I just stay with centos?
Do I tell my hosting provider to just install the OS and give me ssh action and that's it? Don't install any control panels or any other stuff? I want one domain and one subdomain on it though and ftp action.
I want to build one File Server looks like Rapidshare / Megaupload, but this only for internal purposes only, i will put 5 TB only for the harddrive, can you guys recommend the hardware specs for me.
I have a server running windows server 2003, IIS. I have a password protected folder where i have a coldfusion script. I'm trying to write a Batch Script that will run the coldfusion script but the Password Protected folder is giving me problems. I'm using Integrated Authentication on that folder.
anyone know how to write a .BAT script that will load the script and also pass the u/p to that folder so the script can be run?
